US 20060251418 A1
A bit-rate-transparent electrical space-division switching matrix is employed in an optical cross-connect and the input/output stage is constructed from simple, broadband optical receivers and transmitters. Since the switching matrix operates in unclocked manner, i.e. its switching function is not based on internal bit timing and frame timing, arbitrary signals can be switched though transparently at almost any bit rate, independently of the protocol-type being used. The inputs and outputs likewise operate fully independently of bit rate and protocol, since they only implement an O/E conversion or O/E conversion. By virtue of this structure, a simply constructed but extremely powerful optical cross-connect is created that can be employed equally for all types of optical signals within the stipulated wavelength-range. As a result, connections can be switched on the physical layer of an optical network, whereby the optical network can be utilised equally by various services with different transmission protocols.
1. A network element for switching optical communications signals in an optical transport network on the physical layer, comprising:
a number of broadband optical inputs, each for receiving an optical communications signal with a single wavelength channel, each optical input comprising an O/E converter for converting the received optical communications signal into an electrical communications signal;
a number of optical outputs, each for transmitting an optical communications signal with a single wavelength channel, each optical output comprising an E/O converter for converting an electrical communications signal into the optical communications signal to be transmitted;
a transparent space switching matrix operating in an unclocked manner to transparently switch signals of different types, said switching matrix being electrically connected on the input side to the optical inputs and on the output side to the optical outputs, for interconnection of the electrical communications signals from the optical inputs to the optical outputs;
a monitor adapted to monitor signals of one or more predefined types, said monitor being connected to the switching matrix for receiving an electrical communications signal from the switching matrix, and monitoring this communications signal for transmission errors; and
a controller controlling said switching matrix such that only signals of any of the predefined types are switched to the monitor while signals of other types are switched transparently to the respective outputs.
2. A network element according to
3. A network element according to
4. A network element according to
5. A network element according to
6. A network element according to
7. A network element according to
8. A network element according to
9. A network element according to
10. A network element according to
11. A network element according to
12. A network element according to
13. A network element according to
14. A network element according to
15. A network element according to
16. A network element according to
17. A network element according to
This is a continuation of application Ser. No. 10/101,958 filed on Mar. 21, 2002, the disclosure of which is incorporated herein by reference.
The invention relates to an optical cross-connect for switching optical communications signals on the physical layer in an optical transport network.
For the transmission of data in transport networks, nowadays systems are mainly employed that operate in accordance with the recommendations of ITU-T for SDH (Synchronous Digital Hierarchy) and SONET (Synchronous Optical Network). In these systems, payload-data signals to be transmitted are packeted into multiplex units and interleaved in accordance with a multiplex rule to form a time-division-multiplexed communications signal. This is then transmitted as an optical communications signal via optical fibers of the network, whereby in interposed network elements of the transport network the communications signals are disassembled into the individual multiplex units contained therein and can be interleaved again in a new configuration. Connections in an SDH-based or SONET-based transport network are switched in this way. An overview of these systems is presented, for example, in the article entitled “SONET 101” produced by Nortel Networks, which can be downloaded on the Internet from www.nortel.com/broadband/pdf/sonet 101.pdf.
In addition, optical networks are also increasingly finding application in which wavelength-multiplexed optical communications signals are transmitted. Such wavelength-multiplexed signals consist of several optical channels which are each represented by one wavelength. As a result, it is possible for several optical signals to be transmitted in parallel via a single optical waveguide. For such optical networks, network elements are required that are capable of decoupling individual wavelengths from and inserting them into wavelength-multiplexed communications signal (add/drop multiplexers). In addition, network elements are required that are able to disassemble such wavelength-multiplexed communications signals into the individual wavelengths contained therein and multiplex them again in a new configuration, in order to switch connections in the optical network (cross-connects). The realisation of the network elements that have been described is technically very elaborate, and the development of these devices is currently still in its initial stages. A standard for the optical channels is currently still being developed. For instance, a new multiplex hierarchy with the designation “optical channel (OCh)” is currently being discussed. This new multiplex hierarchy is intended to have multiplex levels with bit rates of 2.66 Gbit/sec and also multiples thereof (factors of four), namely 10.7x Gbit/sec and 43.x Gbit/sec, and is provided for optical transmission of information using wavelength-division multiplexing (WDM).
Furthermore, so-called optical cross-connects are currently being developed which are intended to switch optical communications signals of arbitrary format such as SONET, ATM, IP. They contain a central space-division switching matrix which is intended to be transparent in respect of the communications signals to be switched. An example of such an optical cross-connect is specified in the article entitled “Cost Effective Optical Networks: The Role of Optical Cross-Connects” by Charles A. Brackett, which can be downloaded on the Internet from the site www.tellium.com. This optical cross-connect operates with a SONET-based, clocked electrical space-division switching matrix which operates with the frame timing of OC-48. Full transparency in respect of non-SONET-based communications signals is therefore not guaranteed.
A further example is the 20000 Series wavelength router produced by Monterey. In a brochure relating to this product which has been published on the Internet it is specified that said router consists of a subsystem designated as a “switch core” and of I/O cards that support OC-48/STM-16 signals and that are connected to the switch core via short-range optical interfaces (Ultra-Short-Reach-Optics™). The switch core has a switching capacity of 256 OC-48 or 64 OC-192 equivalents. Consequently it is likewise based on the frame-timing rate of SONET and is therefore not fully transparent in respect of all signal-types.
One object of the present invention therefore consists in specifying an optical cross-connect that operates independently of bit rate and independently of protocol and additionally guarantees transmission functions for at least one protocol-type. A further object of the invention consists in specifying a method for bit-rate-independent and protocol-independent switching of optical communications signals.
The object is achieved by an optical cross-connect which has broadband optical inputs, optical outputs, a transparent space switching matrix (S) which is electrically connected on the input side to the optical inputs and on the output side to the optical outputs, and a monitoring device which is connected to at least two matrix ports in order to receive an electrical communications signal from the switching matrix, to monitor this communications signal for transmission errors and to switch the communications signal back to the switching matrix.
An advantage of the cross-connect according to the invention consists in the fact that it permits a remotely controllable interconnection of optical communications signals of arbitrary signal format via a central network-management system.
Further advantages consist in the fact that the cross-connect according to the invention affords cost-savings in production and operation. In particular, the input/output interfaces are extremely cost-effective by reason of their simple structure. In addition, the optical cross-connect can be employed extremely flexibly in all existing and future networks and is open to all existing protocol-types. Current semiconductor technology permits use up to bit rates of 20 Gbit/sec.
It is a particular advantage to connect a monitoring device that is adapted to a particular protocol-type to the switching matrix, so that signals of this protocol-type can be conducted from the switching matrix to the monitoring device and can be conducted back again from the latter to the switching matrix. As a result, a monitoring of communications signals of this protocol-type is made possible without impairing the transparency and functionality of the cross-connect. By virtue of the monitoring device, a protective function can be set up and a predetermined signal quality can be guaranteed for communications signals of this protocol-type. A particularly advantageous configuration of this monitoring device monitors the B1 byte and/or the J0 byte of SDH or SONET signals. As a result, the optical cross-connect can be integrated into existing SDH systems or SONET systems.
The cross-connect according to the invention can be employed in WDM networks together with a wavelength-division multiplexer. In this case it is advantageous that typical topics in the WDL layer such as wavelength allocation and wavelength conversion are decoupled from the switching function in an optical network.
The invention will be elucidated in the following with reference to an exemplary embodiment on the basis of
The optical cross-connect OCX represented in
The optical inputs each possess an O/E converter which converts an optical communications signal into an electrical communications signal. Analogously, the optical outputs possess an E/O converter which converts an electrical communications signal from the switching matrix S into an optical communications signal to be transmitted.
The switching matrix S is a transparent space-division switching matrix which is electrically connected on the input side to the optical inputs and on the output side to the optical outputs of the input/output stage I/O and which optionally interconnects the electrical communications signals from the optical inputs to the optical outputs under supervision of the control circuit CS. The switching matrix operates advantageously in unclocked manner. Similar unclocked switching matrices have hitherto been used in IP routers at a distinctly lower bit rate. Based on the latest Si/Ge-BiCMOS semiconductor technology, the switching matrix S can easily be designed for a maximum transmission rate of the order of 12 Gbit/sec. Laboratory experiments show that such a switching matrix consisting of Si/Ge-BiCMOS semiconductors can even be employed up to a transmission rate of 20 Gbit/sec.
A fundamental idea of the invention consists in employing a bit-rate-transparent switching matrix in an optical cross-connect and in constructing the input/output stage from simple, broadband optical receivers and transmitters. A monitoring device is connected to the switching matrix, so that communications signals can be looped from the switching matrix to the monitoring device and back to the switching matrix, in order to implement transmission-system monitoring functions for communications signals of at least one predetermined protocol-type.
The switching matrix operates advantageously in unclocked manner, i.e. its switching function is not based on internal bit timing and frame timing. As a result, arbitrary signals can be switched through transparently at almost any bit rate, independently of the protocol-type being used. The inputs and outputs likewise operate fully independently of bit rate and protocol, since they only implement an O/E conversion or O/E conversion. For instance, very simple O/E converters consisting, in the simplest case, of a single photodiode are employed in the receivers. In addition, a bit-rate-independent signal processing (signal recovering) of the electrical signal can be performed, so that the resulting digital signal exhibits defined level-states with steep edge waveforms. This can be implemented with a Schmitt trigger, for example.
Correspondingly simple optical transmitting elements such as, for example, directly modulated broadband semiconductor lasers are employed in the transmitters. This takes account of the fact that the inputs and outputs are each provided for the purpose of receiving and transmitting, respectively, an optical communications signal with only a single optical channel within the wavelength-range of 1300 nm or 1500 nm, in which, however, in contrast with WDM systems, the exact wavelength plays no role. The function of this cross-connect is limited only by the maximum bit rate of approximately 12 or even 20 Gbit/sec which can still be switched through transparently by the switching matrix on the basis of the bandpass response which is characteristic of all semiconductors.
By virtue of this structure, a simply constructed but extremely powerful optical cross-connect is created that can be employed equally for all types of optical signals within the stipulated wavelength-range. As a result, connections can be switched on the physical layer of an optical network, whereby the optical network can be utilised equally by various services with different transmission protocols, such as IP, ATM or SDH, for example.
In G.707 (1996), Chapter 22.214.171.124, one byte in the overhead of STM-N frames is reserved for a checksum. This byte is designated as the B1 byte; the checksum is designated by BIP-8 (Bit Interleaved Parity 8). The BIP-8 code is a checksum with even parity which is calculated over all the bits of the preceding frame after the scrambling thereof and is written to the B1 byte of the current frame before the latter is scrambled.
Such a BIP-X code is defined for a process for error-monitoring. It is calculated by the transmitting device over a predetermined part of the signal to be transmitted, specifically in such a way that the first bit of the code generates an even parity over all the first bits of all x-bit sequences of the signal part that is covered, the second bit of the code generates an even parity over all the second bits of all x-bit sequences and so on. Even parity means that the BIP-X bits are assigned in such a way that an even number of (logical) ones arises in the monitored partition of the signal. The monitored partition comprises all the bits in the same bit position within the x-bit sequences in the part of the signal that is covered, said part of the signal containing the BIP-X code itself.
In the exemplary embodiment the monitoring device MON is designed in such a way that it calculates the BIP-8 code for each STM-N frame in a communications signal, reads out the BIP-8 code from the B1 byte of the following frame, and compares the two with one another. If a discrepancy is established, an error has occurred.
Now the monitoring device MON determines various parameters that characterise the quality of the transmission. In this connection, the errors that have occurred and that have been established on the basis of the BIP-8 calculation are determined and counted over predetermined time-intervals—typically 15 min and 24 h. Typical parameters for this purpose are: “background block errors”, “errored seconds” and “severely errored seconds”. An “errored second” is present if errors are present in a one-second interval and, however, less than 30% of the frames are defective. A “severely errored second” is present if more than 30% of the received frames are defective in a one-second interval. The parameters are added over 15 min and over 24 h. These values 19, represent monitoring data. They enable a check of the actual quality of the transmission and a protection circuit, inasmuch as a change-over to a redundant back-up line is undertaken if a predetermined transmission quality is fallen short of. This will be dealt with in more detail at a later point in connection with
Additionally, or alternatively, it is advantageous also to implement in the monitoring device a second monitoring function for SDH signals. According to ITU-T G.707, Chapter 126.96.36.199, a seven-bit checksum CRC-7 (Cyclic Redundancy Check) is contained in the so-called J0 byte. The J0 bytes of 16 consecutive STM-N frames constitute one 16-byte frame which contains the so-called Section Access Point Identifier (API). The API is constantly repeated, so that the receiving side can check whether the connection to the intended transmitter still exists. The first byte of the API contains, at bit positions 2 to 8, the checksum that is calculated from the preceding STM-N frame. The monitoring device MON reads the J0 byte from the STM-N frames of a communications signal. On the basis of the J0 byte the continuity of the intended connection is checked and is tested by means of the checksum CRC-7 as to whether the preceding frame was error-free.
The monitoring device is connected to the switching matrix S via at least two ports. If a communications signal is now to be monitored, said communications signal is switched from the input/output stage I/O to the monitoring device MON via the switching matrix. Said monitoring device performs the monitoring functions and conducts the unchanged communications signal back again to the switching matrix S. From the switching matrix S the communications signal is then switched to the stipulated output of the input/output stage I/O.
Consequently a communications signal to be monitored is switched twice by the switching matrix S: a first time to the monitoring device, and a second time to the intended output of the cross-connect. As a result, a monitoring of the quality and the path of communications signals in the optical network is made possible without the transparency and functionality of the cross-connect being impaired.
Since the monitoring device performs protocol-specific functions, it has to be adapted to a particular protocol-type or to a group of different protocol-types that is able to distinguish it and handle it appropriately. The signal-monitoring described above may, of course, be employed advantageously not only in combination with a bit-rate-transparent electrical switching matrix, as in the present exemplary embodiment, but in combination with any type of switching matrix.
The control device CS controls the function of the switching matrix S, i.e. it specifies to the switching matrix S which matrix input is to be switched to which matrix output. The control device CS possesses a terminal TMN leading to a central management system which is not represented. This gate terminal (TMN) may, for example, operate on the basis of the SNMP (Simple Network Management Protocol) known from IP routers or even with the CORBA (Common Open Request Broker Architecture) specified by OMG (Object Management Group) or with another protocol that is suitable for network management. The central network-management system has an overall view of the entire network and can decide, on the basis of this information, which connections the individual cross-connects are to switch. This is then communicated to the control device CS of the cross-connect in question via the gate terminal. The control device also collects and stores the monitoring data of the monitoring circuit, so that said monitoring data can be called up on demand by the central management system via the gate terminal. Alarms can also be signalled to the central management system by this method. The gate terminal may, of course, also be a channel in one of the communications signals.
A portion of a network with four optical cross-connects OCX1-OCX4 is represented in
In the same way, the communications signal is then switched from the second cross-connect OCX2 to the third cross-connect OCX3 and from there to the fourth cross-connect OCX4. The fourth cross-connect OCX4 then switches the communications signal to an output of its input/output stage I/O that is connected to the data sink S2. A monitoring, of the type described above, is implemented in all four cross-connects OCX1-OCX4. The data source and the data sink may, for example, each be a line multiplexer which generates from several tributary signals an SDH signal consisting of STM-4 frames. The optical connection through the network has been described only in the direction from S1 to S2 but may, of course, also be bidirectional.
By virtue of the monitoring in the four optical cross-connects OCS1-OCS4, a predetermined quality of transmission is ensured. In case one of the cross-connects establishes that errors are occurring frequently and the quality of the transmission does not satisfy the preset requirements, an alarm message to the central network-management system can be generated, for example.
The cross-connect according to the invention may also be employed in optical networks with wavelength-division multiplexing. To this end, several input/output ports of the cross-connect are connected to an optical wavelength-division multiplexer. This is represented in
In this exemplary embodiment the wavelength-division multiplexer is a line multiplexer. But one or more ports of an add/drop multiplexer may also be connected to the optical cross-connect in exactly the same way. This is shown in